Available models: Qwen3.5 397B A17B (Qwen) — $0.54/M input, $3.40/M output · Qwen3.5 122B A10B (Qwen) — $0.29/M input, $2.90/M output · Qwen3.5 35B A3B (Qwen) — $0.22/M input, $2.20/M output · Qwen3.5 9B (Qwen) — $0.04/M input, $0.20/M output · Qwen3.5 4B (Qwen) — $0.03/M input, $0.15/M output · Qwen3.5 2B (Qwen) — $0.02/M input, $0.10/M output · Qwen3 235B A22B Instruct 2507 (Qwen) — $0.07/M input, $0.10/M output · Qwen3 Next 80B A3B (Qwen) — $0.09/M input, $1.10/M output · Qwen3 Coder 480B A35B Turbo (Qwen) — $0.22/M input, $1.00/M output · Qwen3 Coder 480B A35B (Qwen) — $0.40/M input, $1.60/M output · Qwen3 235B A22B Thinking 2507 (Qwen) — $0.23/M input, $2.30/M output · Qwen3 VL 235B A22B (Qwen) — $0.20/M input, $0.88/M output · Qwen3 VL 30B A3B (Qwen) — $0.15/M input, $0.60/M output · DeepSeek V3.2 (DeepSeek) — $0.26/M input, $0.38/M output · DeepSeek V3.1 (DeepSeek) — $0.21/M input, $0.79/M output · DeepSeek R1 0528 (DeepSeek) — $0.50/M input, $2.15/M output · DeepSeek R1 0528 Turbo (DeepSeek) — $1.00/M input, $3.00/M output · DeepSeek R1 Distill Llama 70B (DeepSeek) — $0.70/M input, $0.80/M output · DeepSeek OCR (DeepSeek) — $0.03/M input, $0.10/M output · Llama 3.1 8B Turbo (Meta) — $0.02/M input, $0.03/M output · Llama 3.2 11B Vision (Meta) — $0.05/M input, $0.05/M output · Llama 4 Scout 17B 16E (Meta) — $0.08/M input, $0.30/M output · Llama 3.3 70B Turbo (Meta) — $0.10/M input, $0.32/M output · Llama 4 Maverick 17B 128E Instruct FP8 (Meta) — $0.15/M input, $0.60/M output · Llama Guard 4 12B (Meta) — $0.18/M input, $0.18/M output · Kimi K2.5 (Moonshot) — $0.45/M input, $2.25/M output · Kimi K2.5 Turbo (Moonshot) — $0.60/M input, $3.00/M output · Kimi K2 Thinking (Moonshot) — $0.47/M input, $2.00/M output · MiniMax M2.5 (MiniMax) — $0.27/M input, $0.95/M output · GLM 5 (Zhipu) — $0.80/M input, $2.56/M output · GLM 4.7 Flash (Zhipu) — $0.06/M input, $0.40/M output · GLM 4.6V (Zhipu) — $0.30/M input, $0.90/M output · gpt oss 120b (OpenAI) — $0.04/M input, $0.19/M output · gpt oss 20b (OpenAI) — $0.03/M input, $0.14/M output · gemma 3 27b it (Google) — $0.08/M input, $0.16/M output · gemma 3 12b it (Google) — $0.04/M input, $0.13/M output · Mistral Small 3.2 24B Instruct 2506 (Mistral) — $0.07/M input, $0.20/M output · NVIDIA Nemotron Nano 9B v2 (NVIDIA) — $0.04/M input, $0.16/M output · Llama 3.3 Nemotron Super 49B v1.5 (NVIDIA) — $0.10/M input, $0.40/M output · phi 4 (Microsoft) — $0.07/M input, $0.14/M output